This invention relates to apparatus and process for analyzing a liquid to be tested such as, for example, a urine sample collected from a living body. The liquid flows in a flattened sheet, surrounded by a sheath liquid. The orientations of respective ingredients in the liquid to be tested are placed in a condition that permits the classification and counting of the ingredients through image processing. More specifically, during the measurement, the classification and counting are carried out by changing the magnification of the optical system.
Microscopic examination of urine sedimentation is a well-known clinical examination technique that continues in common use. This examination requires only a very simple collection of urine from a patient. An examination of sedimentation in the urine, can help determine the conditions of kidneys, urinary and genital organs, since urine contains red blood corpuscles from capillary vessel bleeding, white blood corpuscles from the vessel, epithelium of kidney or urinary/genital organs, columnal kidney tubules and microorganisms associated with infection.
In a conventional examination, components in the urine are separated in a centrifuge. The sediment is placed on a microscope slide to make a sample. Then, the ingredients of the sample are examined and classified by microscopic observation.
Conventional examination suffers from the disadvantage that some ingredients of the urine are damaged during centrifugal separation. The accuracy of determining the concentration of a component is complicated by the process of centrifugal separation. Further, microscopic observation needs a very precise classification of respective ingredients, and the number of the ingredients to be observed is small. In addition, the ingredients are scattered irregularly within the sample. This makes the process very burdensome for examining technicians. As a result, conventional examination may produce errors in the analysis of the urine.
A more accurate and less labor-intensive technique employs an automatic urine analyzing device. A liquid to be tested, surrounded by an outer layer of a sheath liquid, is transformed into a very flattened flowing sheet of liquid. The flattened sheet of liquid is photographed by a video camera. The resulting static image is analyzed by image processing techniques to classify and count the ingredients in the liquid to be tested.
Microscopic examination of such a flattened sheath flow method must meet the following two (2) requirements:
(1) Flattened ingredients such as red blood corpuscles must be oriented in a certain direction. This aims at obtaining the image of the ingredients.
(2) The liquid to be tested must be transformed into a flattened flow liquid. The thickness of the flattened flow liquid is less than the depth of field of the video camera. This produces a very well-focused image. Also, by flattening the liquid to be tested, the particles to be detected are spread widely across the photographic image.
Examples of the foregoing flattened sheath flow method are disclosed in Japanese Patent Publication 57-500995 and U.S. Pat. No. 4,338,024. In order to photograph the static image of the ingredients in the flattened sheath flow, a strobe light having a short emitting time or a pulse laser light is irradiated through the thin cross section of the flattened flow. The image is photographed through the objective lens of a video camera.
The kinds and sizes of ingredients in a urine sample vary. For example, the diameter of a red blood corpuscle is about 10 micrometers. The diameter of the epithelium is a little larger, i.e., several tens of micrometers. In some cases, a columnal tubule is from several hundred micrometers to 1 or 2 millimeters in length. It is impossible to photograph and analyze respective ingredients having such a range of sizes at the same magnification.
Therefore, the magnification must be changed during a measurement process. Smaller ingredients are photographed at a high magnification, and larger ingredients are photographed at low magnification.
However, when the magnification of the lens is changed, the depth of field of the lens is also changed. At high magnification, the depth of field is reduced. In the flat-sheath flow method, if the flattened stream is thin enough, focussing can be accomplished at both high and low magnification. However, using the same thickness for both magnifications has a drawback. At low magnification, a stream that is thin enough to maintain focus at high magnification, is too thin to contain enough of the larger particles that are the target at such low magnification. The small number of particles degrades measurement accuracy.